Proceedings of 2015 IFToMM Workshop on History of Mechanism and Machine Science
May 26-28, 2015, St-Petersburg, Russia
HISTORY OF BIPED ROBOTS. PROBLEMS.SOLUTIONS
Anastasiya Borina
candidate
Saint-Petersburg Polytechnical University
Russia, Saint-Petersburg
Тел.: +7(905)285-5112
E-mail: [email protected]
Valeriy Tereshin
docent, Cand.Tech.Sci.
Saint-Petersburg Polytechnical University
Russia, Saint-Petersburg
Тел.: +7(812)297-4845,
E-mail: [email protected]
Abstract: This paper presents the history of biped robots
creating from ancient time to today. The fist devices simulated
human and animal motions, but had no "sense organs". Today
many modern interesting walking devices are created. The main
problem in this field is modeling and dynamic control.
Keywords: walking robot, history of biped robots, movement
control, dynamic stability
Introduction
Walking robots are the most wanted, important and
interesting ones and their stability is the most crucial problem
these days. Two main types of bipedal walking are present in the
literature: static and dynamic [1]. The static walking is always
stable, but such robots have low speed and big weight [2].
Dynamic walking is considered to be human like. As opposed
static, dynamic robots are faster, more maneuverable and
dexterous. The feet of such robot are very small, like dots, soit is
impossible to be stable for them. Dynamic walking means to
constantly fall, but to bring forward the swing leg in time to
prevent tilting over.The control system must provide stability of
walking and its efficiency [6],[21]. So it is the main problem at
creation of the modern walking robots.
Researches of dynamic walking are connected with names of
P. L. Chebyshev (1824-1894), I.I. Artobolevsky (1905-1977),
A.E. Kobrinsky (1915-1992), Yu. I. Neimark (1920-2011), D.E.
Okhotsimsky (1921-2005), Yu.V. Bolotin(1926-2008), V. V.
Beletsky (1930-), M. Vukobratovich (1931-), E.A. Devyanin
(1931-2002), A. K. Platonov (1931-), A.M. Formalsky (1938-),
etc. The first walking robots were used for entertainment. More
advanced walking robots were developed with the various
branches of technology and science [7]. They turned into devices
and mechanisms which could exempt people from a hazardous
to health work, in the conditions of the raised radiation, high or
low temperature, in hard-to-reach areas, and also for
rehabilitation of people with the limited possibilities [14].
Figure 1. Mechanical knight of Leonardo da Vinci
The French mechanic Jacques de Vocanson in 1738 built the
androids which made him the world famous. The robot-flutist
was tall like the adult and was able to hold a flute at lips.
Blowing and pressing flute valves, the machine played 11
various melodies. Another the most known invention of
Vocanson is the Mechanical duck (Figure 2).
Figure. 2. Mechanical duck of Jacques de Vocanson
1 The fist walking robots
The robot is the Czech word. In 1921 Czech writer Karl Capek
introduce the play "R.U.R" Rossum's Universal Robots, where
the androids revolted against people. These human-like
machines are called Robots based on the Czech work robota
meaning labor [17]. In 1495 Leonardo da Vinci performed the
first the detailed project of walking robot, it was able to move
hands and turn the head (Figure 1).
It consisted of 1000 details. The duck was covered with the real
feathers, was able to walk, move wings, grunt, drink water, peck
grain and, after milling it inside, crap on a floor [9]. In 1865
Johnny Breinerd built the Steam Man (Figure 3).
In 1963 in Cornell University's (New York) laboratory of
aeronautics N. Mayzen developed passive exoskeleton
(Figure 5) which registered the movements of all links of a body,
except a neck and fingers.
Figure 3. Steam Man by Breinerd
Breinerd's mechanism was three-meter high, it easy for it to
pull acart with five passengers. Instead of a hat, the Steam Man
had a flue pipe with the black smoke. The Steam Man was able
to move with a speed up to 30 miles an hour.
All mentioned inventions were interesting, but these devices
had no mechanisms which could inform them where and when it
was necessary to put a foot [15]. After discovery of uranium in
1896 and its harmful effects on an organism it became necessary
to create the robots which could replace people during work with
radioactive materials. So it was necessary to solve one of the
most difficult scientific tasks – how to control the walking robot.
2 The development of robotics
In 1878 the Russian mathematician Chebyshev presented a
model for a locomotion system at the World Fair in Paris
(Figure 4).
Figure 5. Passive exoskeleton
The obtained data represented interest for development of an
active exoskeleton, which was able to support at least a portion
of the load of a user's body.
In 1966 Ichiro Kato from the Waseda University in Tokyo,
started his work on biped robots. His first robot WL-1 is shown
in Figure 6.
Figure 4.Chebyshev multiple-bar walking mechanism
Chebyshev's mechanism was the first walking machine and
survived as a landmark in lokomotion research.
During the period from 1900 to 1940 not so many walking
mechanisms were invented because of the Russo-Japanese War
and the World War I. The most interesting of them was a robot
which lectured at the «Century of Progress» exhibition in
Chicago in 1933. Its breast and stomach were transparent and
robot showed a gullet, a stomach, intestines and a liver and
explained a structure of internals. The first Russian robot
android B2M was created in 1936 by the schooler Vadim
Matskevich and in 1937 he was awarded the diploma of the
World Fair in Paris. In 1939 at the World Fair in New York
«Westinghouse Electric Corp».presented the mechanical
humanoid robot «Elektro» and the robot-dog «Sparko». Elektro
was 136 kg, was able towalk, talk and smoke.
In the 1960th computer technologies and facilities began to
develop so it became possible to create the controlled walking
robots [16]. Also in that period the medicine actively developed
andin the USSR, the USA, Yugoslavia, Italy and Germany the
special devices for people who were injured were designed.
Such devices were called exoskeletons.
Figure 6. The walking robot of the Waseda University
WL-3 from 1969 performed static walking. In 1970
WABOT-1 was built, it accepted as the first fully articulated
anthropomorphic biped. The top part and center of gravity of
robot's body moved to the left and to the right due to the hinges.
The program control of walking was applied in this robot.The
walking was considered to consist of some consecutive
movements [19]. The program of walking was stored in the
block of memory and could be changed. Process of walking was
synchronized by the impulses given in certain intervals of time.
Each hinge was equipped with the potentiometer which signals
served as feedback. Robot turned by manual switch. Its speed
was slow and the walking was static.
In the 1970th M. Vukobratovich and the institute of
automatic equipment and telecommunications of M. Pupin in
Belgrade worked on creation of the biped device— the active
exoskeleton. It consisted of a metal framework, its links and
hinges were like the bones and joints of the person [18].
Exoskeleton was equipped with pneumatic actuators,
substituting of muscles, and system of the sensors signaling
about position of joints. It was put on the person. The device was
supplied with the portable program device providing the stable
anthropomorphous walking [4], [10]. Developers investigated
biomechanical and physiological properties of movements.
Synthesis of anthropomorphous walking was the original. Some
coordinates was set like a nominal law of the movement, and
values of the remained coordinates were calculated from the
dynamic links. The three-level control system was created for
this robot. Ideas of the Soviet physiologist prof. N. A. Bernstein
made a great impact on formation of such approach to biped
control.
3 Robots today
There are not so many modern devices deserving attention
today. For example «Walking assist device» by Honda
(Figure 7).
4 The control problem of bipedal robots
There are many modeling and control problems but we will
mainly focus on a stabilization of the biped walking in the field
of specified movement. In general, a bipedal locomotion system
consists of several members that are interconnected with
actuated joints. Its point of support changes discretely [11], [13].
The parameters of control are the time of the beginning of the
next step and coordinates of a reference point [3]. So to stabilize
walking it is necessary to define «when and where to put a foot».
Feedback is based on the equations of the ideal mechanism the turned pendulum [20]. The analysis of its movement allows
to solve a problem "when and where to put a foot". In this case
this problem has solution set. For modeling a control system it is
necessary to work out the nonlinear differential equations, let
initial conditions are equal to final conditions from the previous
step [2]. The Figure 9 shows the structure of the biped and
coordinates used to describe the configuration of the system.
Right
hip joint
Left
hip joint
Body
Оr2
Right leg
Оr1
Z
Fr
C.G.
Оl2
Left leg
Оl1
Y
Figure 7. Walking assist device by Honda
Walking assist device is an exoskeleton, it is intended, as a
rule, for increase human efforts. Walking assist device can be
used for making hard work, such as loading in places difficult of
approach for wheel transport [8], for repair work, rescue
operations. Besides, exoskeleton can serve for rehabilitation of
the injured people.
It is necessary to mention the walking chair WL-16 by the
Waseda (Figure 8).
O
Fl
X
Figure 9. Kinematic model of a biped robot
This paper proposes the method of biped horizontal walking
control of flat machine by defining - the time of the end of the
current step, when
, i-step number, and place of
touchdown at the beginning of the next step. The changing the
length of the leg provides horizontal movement of the device [5].
Error in emplacement is defined as a difference between point
coordinate at vertical position of the leg
and its program
value
in the same point of time.
Let the error
accepts very big negative values and
maximum parameter of a step
are defined by design, then
the nonlinear function of control is equal to zero when the error
is absent, and the nonlinear function is equal to admissible value
when the error is maximum. For example
is in range from
to
. .Coordinates of the center of gravity at the
beginning of the step can be represented as
Figure 8. The walking chair WL-16
The walking chair WL-16 was created as the alternative to
wheel chair which is not adapted for walking up and down the
stairs. Mechanisms of feet have 6DOF.
Due to the parallel connections of drives the walking chair
has high mechanical rigidity and precision of movements that
increases its loading capacity and the speed of movement [12].
where
and
– coefficients of feedback by error and its
speed, "·" – time derivative. Define speed
where
; L – height of center of gravity;
– the
acceleration of free fall. Define the coordinates of center of
gravity of the next step as:
4.
From this point of view when the device approach closer to
the purpose decreases from
to
, and
from
zero to
.
Let
. The Figure 10 and
Figure 11 present the results of the numerical solutions of the
equations (1) - (4).
2,5
5.
6.
(T),m/s
2,0
7.
1,5
1,0
8.
0,5
z(T), m
0,0
-0,5
0,0
0,5
1,0
1,5
2,0
2,5
3,0
9.
Figure 10.Simulated response
of the accelerated walking
3,0
z(T), m
10.
2,5
2,0
11.
1,5
1,0
0,5
12.
t,s
0,0
0,0
0,5
1,0
1,5
2,0
2,5
3,0
Figure 11. Temporary dependence
of the accelerated walking
5 Conclusions
(1) From aforesaid it is clear that dynamic biped walking device
must be equipped with sensors of absolute linear and angular
coordinates, and also sensors of position of kinematic links.
(2) The control system has the hierarchical structure. The lowest
level of control is organized like the watching drives and
provides desirable position of the device on all possible
coordinates.The horizontal coordinates and rotation round
the vertical are stabilized at the top level of control
according to movement of the device.
(3) In a feedback the element of comparison is the ideal
mechanism, it is described by two equations of the turned
pendulum for coordinates x and z synchronized at the end of
the step.
(4) Stabilization of rotation round a vertical is carried out when
both feet are on the ground.
13.
14.
15.
16.
17.
18.
19.
References
1. Automatica. (1999). A special issue on hybrid systems.
Automatica, 35(3), 374–535.
2. Borina A, Tereshin V. (2015). Control of biped walking
robot using equations. Advances in Mechanical Engineering,
Lecture Notes in Mechanical Engineering (pp. 23–31)
Switzerland: Springer.
3. Borina А.P., Теreshin V.А. Reshenie zadachi o
prostranstvennom dvizhenii staticheski neustoichivogo
shagaushego apparata. Sovremennoe mashinostroenie.
Nauks i obrazovanie: materialy nauchno-prakticheskoi
konferensii s mezhunarodnym uchastiem. 20 – 21 iunya
20.
21.
2013 godа, Sankt-Peterburg. – SPb.: Izd-vo Politehn. un-tа,
2013, S. 631-641. (rus)
Borina А.P., Теreshin V.А. Prostranstvennoe dvizhenie
shagaushego apparata pod deistviem upravlaushih
vozdeistvii so storony nogi. Sovremennoe mashinostroenie.
Nauks i obrazovanie. 2012. № 2. S. 177-188. (rus)
Brogliato, B. (2003). Some perspectives on the analysis and
control of complementarity-systems. IEEE Transactions on
Automatic Control, 48(6), 918–935.
Brogliato, B., Niculescu, S. I., & Monteiro-Marques, M.
(2000). On the tracking control of a class of
complementarity- slackness hybrid mechanical systems.
Systems and Control Letters, 39, 255–266.
Brogliato, B., ten Dam, A. A., Paoli, L., Genot, F., &
Abadie, M. (2002). Numerical simulation of finite
dimensional multibody nonsmooth mechanical systems.
ASME Applied Mechanics Reviews, 55, 107–150.
Ceccarelli M., Li Hui, Carbone Giuseppe and Huang Qiang.
Conceptual Kinematic Design and Performance Evaluation
of a Chameleon-like Service Robot for Space Stations. Int J
Adv Robot Syst, 2015, 12:17. doi: 10.5772/60203.
Ceccarelli M., Li H., Carbone G., Huang Q., 2012,
Conceptual Design and Characterization by Simulation of a
Chameleon-like Robot for Outdoor Space Service,
Proceedings of RAAD 2012 international Workshop on
Robotics in Ale-Adria-Danube Region, Napoli, Edizioni
Scientifiche e Artistiche, 334-341.
Ceccarelli M., 2011, Problems and issues for service robots
in new applications, International Journal of Social Robotics:
Volume 3, Issue 3 (2011), 299-312.
Ceccarelli M. (ed.), 2012, Service Robots and Robotics:
Design and Application, Engineering Science Reference, IGI
Global, Hershey
Chevallereau, C. (2003). Time-scaling control of an
underactuated biped robot. IEEE Transactions on Robotics
and Automation, 19(2), 362–368.
Chevallereau, C., & Aoustin, Y. (2001). Optimal reference
trajectories for walking and running of a biped robot.
Robotica, 19, 557–569.
Collins, S., Wisse, M., & Ruina, A. (2001). A 3-d passive
dynamic walking robot with two legs and knees.
International Journal of Robotics Research, 20(7), 607–615.
Dankowicz, H., Adolfsson, J., & Nordmark, A. (2001). 3D
passive walkers: finding periodic gaits in the presence of
discontinuities. Nonlinear Dynamics, 24, 205–229.
Das, S. L., & Chatterjee, A. (2002). An alternative stability
analysis technique for the simplest walker. Nonlinear
Dynamics, 28(3), 273–284.
Evgrafov A, Egorova O, Rayushkina V. (2013) Modern view
on “History of science” of teaching at technical universities.
New trends in educational activity in the field of mechanism
and machine theory, Springer, 137-145.
Gienger, M., LNoOer, K., & Pfei:er, F. (2003). Practical
aspects of biped locomotion. In B. Siciliono, & P. Dario
(Eds.), Experimental robotics (pp. 95–104). Berlin,
Heidelberg: Springer.
Kar, D. C., Kurien, I. K., & Jayarajan, K. (2003). Gaits and
energetics in terrestrial legged locomotion. Mechanism and
Machine Theory, 38, 355–366.
Lum, H. K., Zribi, M., & Soh, Y. C. (1999). Planning and
control of a biped robot. International Journal of Engineering
Science, 37, 1319–1349. Mabrouk, M.
Saidouni, T., & Bessonnet, G. (2003). Generating globally
optimised sagittal gait cycles of a biped robot. Robotica, 21,
199–210.